PCB fuser for toner transfer etching


We’ve covered the toner transfer method for creating circuit boards before. Instead of using an iron, Rich uses the fuser out of a laser printer to get consistent results. He wired a dimmer to the power supply to regulate the temperature. The fuser has a built in thermistor, so you can read the temp using an ohm meter. For rate control he built a stepper controller that connects to a parallel port. Using a simple program he’s able to control both speed and direction.

[thanks Fredasp]

35 thoughts on “PCB fuser for toner transfer etching

  1. i still prefer the photo etch method. people say its soo much more expensive and stuff, but the only real expensive thing is the lamp, which could cost around 30 bux if you look hard. and it really only takes 1 more chemical (the photo stripper). but honestly, its easier is more consistent and can do much much much finer traces.

  2. Interesting method, certainly looks a lot more reliable than using an iron. I’ll have to check on how laser printers work but maybe it could be possible to print the toner directly on the PCB instead of doing a transfer?

  3. Nice hack, but pretty dangerous. (as the man says) If anyone was to attempt this on a more permanent scale, might be a good idea to make some covers to limit the dangers of the exposed PSU and fuser unit etc.

  4. Great idea.

    I agree it adds some complexity which was the advantage toner transfer had over UV exposure, but look at his example PCB, looks perfect.

    Plus its still ends up cheaper, (if you can pick up an old laser printer for next to nothing).

  5. I was wondering if this would be possible… hell, you can take almost the entire fuser assy. out of the printer. Then it’s just the HVTS (High-voltage Transformer) and the gearing. the HVTS is what makes the quartz/halogen lights turn on (that’s what’s inside the hot-roller, BTW.) and stuff. That’s where the real dangerous stuff is. 1000’s of volts around that area. And once you take the rest of the printer away, you don’t have to worry about the tray and the other mechs getting in the way.

    Oh, and you can adjust the pressure roller to be tighter or looser. it’s just a bit more work. Also, this method would work well for double-sided boards. Just flip, and send it back through the fuser.

  6. Since the laser printer usually works by ionizing the paper/plastic so that the toner sticks to it, it might not be possible to ionize small areas of the conductive copper cladding. It’s worth a try.

    Alternatively you could have it apply the ionization and powedered ink onto a rubber roller and then transfer that onto the PCB. You might be able to find a printer that works something like this, but if not, we’re talking about fabricating practically a whole new machine. Sounds like fun. :)

    If anyone found a laser printer that was easy to convert to a PCB printer, they might be able to make a business out of it.

  7. well moogle, just for proof on concept I sent a 8.5″x11″ piece of aluminum foil through my laser printer and lo and behold, it printed with even better precision than it usually does on paper! To the best of my understanding, the sheet itself isn’t ionized but passes between the charged imaging drum and a corona wire or something. The toner particles are attracted from the drum towards the wire and simply deposit on the sheet before going through the fuser assembly.

    Or maybe it’s just that metal can hold a local electrostatic charge for a short while and the toner sticks to it. I’ll try to get my hands on an old LJIII or something with manual toner density control.

  8. very neat Patrick!
    yes I think the “drum” is electrically charged, not the paper. the only thing that needs modifying is some adjustment for the thickness of the PCB and making the PCB go strait through without any bending. Depending on the construction of the printer, that may or may not be a lot of work. I should have an old A4 laser printer somewhere, may be worth taking a look at :-)

  9. You would also have to add an extra fuser stage or completely bypass the normal thermal fuse in the first fuser because the much higher thermal mass of a PCB would make it considerably harder to heat up.

    Also, getting a good enough alignment when doing two-faced boards will be a challenge.

  10. That’s a relief. I normally don’t throw anything away but I was afraid I only saved the useful parts of the printer. But it’s still in one piece. It’s a Acer 7278 A4 laser printer. It even has an expansion slot for additional fonts :-) (well before the USB-age as well)
    It will need some modifications to take solid materials instead of paper, but the framework is all steel and it has a manual knob to set the brightness.

  11. joelanders:

    you have to scratch it with a nail to be able to remove it but it was printed on the mirror side of the foil, which is a very uniform surface and offers little adherence compared to a “brushed” PCB board (which is used for the usual toner transfer method).

    Also, the ink cartridge in the laser printer I used (oldie Laserjet 5L) was almost empty, so there were a few paler areas. I didn’t want to try it on the printer I’m currently using in case it shorted something inside.

  12. I also wouldn’t touch the tin-foil while it’s partly inside the printer. It conducts a lot more than normal paper…
    I also found a printer cable. What are the changes of getting an old laser printer running under Linux these days? (redhat 9)

  13. Ok, I had a non-standard printer cable which happened to fit the 25-pin _serial_ port of the printer… With the right cable it does print now, but it just prints the postscript code as text :-( I tried several, but I don’t think I’m going to find a working driver. Somewhere on the label the text “acer” is printed, but I think it’s a nameless printer, and it’s very old. The model number (7278) does not lead to anything useful. I don’t think it’s going to work.
    It looks a bit like this, but the paper comes out at the top:

  14. Ha! got more data on Acer Acerlaser printers, they apparently use Minolta print engines. Check for similarities between your printer and the Epson Actionlaser 1000, Epson Actionlaser 1500 or Citizen ProLaser 6000.

    According to it should be compatible with the HP Laserjet+ / Laserjet IIP. Will add more info if I find anything.

  15. YES! I already tried HP laserjet without luck, but HP laserjet-II works :-)
    I will have a look tomorrow if I can mod it for printing to solid surfaces. (i.e: remove most of the paper transport system.)

  16. I just ran the A4 sheet of aluminium foil experiment too. It was very noisy, and it came out very wrinkled, but is sure is printing normally. I printed on the not-so-shiny side, and it sticks fairly well, I have to use my nails to get it of the aluminium foil. I think the aluminium foil doesn’t get hot enough, so that will be a problem with PCB’s. But you could always heat-treat the PCB’s afterwards :-)
    (P.S: don’t try this on a new laser printer. the aluminium foil may come out in pieces (if at all))

  17. This laser printer is a complex piece of kit!
    there are 3 main drive shafts comming form somewhere deep inside and another two clutches. These 5 drive shafts seem to be able to operate independently to control sheet feeding, paper transport, drum rotation and toner mixing.
    The interesting part is that there is a thin wire (marked with high voltage stickers) just under where the paper touches the drum. it seems to pull the toner from the drum to the paper by discharging the drum through the paper. but since it seems to work with aluminium foil, it should work with PCB’s. If I directly connect the wire to the PCB itself then I don’t think the wire under the PCB is needed anymore.
    I will need to make some fairly large modifications to make something solid go through. For starters I will remove the heating thing at the end and make the paper/pcb come out at the back instead of the top.

  18. I wouldn’t actually recommend that. The copper layer being highly conductive, connecting it to the corona wire (the thin HV wire) would either drastically lower its resistance and mess up both the electrostatic field and the HV supply or simply short with the drum and cause a flash discharge, making the toner fall off and probably damaging the drum wrapping.

  19. I was later thinking the same thing. but i can just put it under (or near) the pcb. normally the paper is pressed against the drum and the drum is discharged after that, so I expected that the drum would discharge as soon as the aluminium foil got near it since the foil would distribute the discharging field over it’s whole surface. But the toner does not seem to fall of the drum to early, since it’s still printing sharp on aluminium foil.

    I think I have figured out the drive train:
    -switch detects paper in tray.
    -printing is started.
    -a sheet is pulled in by engaging an electric clutch, which activates the roller at the edge of the tray.
    -the sheet is pushed against a set of rollers that are standing still, until a switch is activated.
    (probably to position the paper correctly.
    -these rollers are then activated by another electric clutch and push the paper along the drum.
    -the heater element pulls the sheet further through, activating a 3rd switch to indicate that the sheet came out correctly (no paper jam)

    The paper transport is far from strait, and won’t work for non-bending PCB’s.
    One option would be to replace the whole paper feeding mechanism with a moving table, mounted on the linear bearings from an old A4 scanner. I can then fake the switches using a timer triggered by the electric clutches, and just use the drive that pushes the sheets along the drum to move the table along the drum.
    that way I can tape the PCB onto the moving table so it will be properly aligned and i can print on small PCB’s.
    It’s a pity that the toner/drum won’t work upside down (the toner would fall of the PCB). otherwise I could make it into a flat-bed laser printer, so it could be used to print on almost anything.
    anyway, I would like to try a piece of PCB first, if needed, by manually pulling it along the drum, but it’s not easy to make the printer work with too many parts removed.
    this will take a bit of time :-)

  20. I had this idea a couple of months ago. I started researching it, and I’m pretty sure nobody has done it before. I was even tempted to file a patent application! Guess there’s not much point in that now!

    I bought an old printer, and started thinking about flattening it out, but I’m too busy with university work at the moment.

    I reckon there could be commercial interest in this – many firms want to be able to do very fast in-house PCB prototyping, and the quickest option is currently a CNC mill, which is quite expensive and is incapable of the resolution required for many modern IC packages.

    If you do a good job, you might end up with something marketable!

    Good luck!

  21. I’m the guy who did the laserjet hack for making PCBs. I just discovered a much better technique. I borrowed a GBC “creative laminator” from my wife’s workplace and it is superb for doing toner transfer PCBs. Let it heat up fully and run the PCB with the toner paper on TOP (the top roller is the heated one) about 5 times in a row to let the heat build up in the board. I’m getting very consistent results on 1/16 FR4 board – almost photographic quality. For thinner or smaller boards tape the PCB & toner sandwich to a thin cardboard carrier to make it a bit thicker and for ease of handing.

    The GBC “creative laminator” is the older blue model. I picked one up at Staples for $47 cdn. Unfortunately the new version of this model has a thinner opening which will not accomodate 1/16″ board :(


  22. Hi all, i hope this discussion is not altogether dead. There is some discussion about using a laser printer to print toner direct to the pcb. Ive been toying withthe idea of using the laser diode assy to print direct on the photosensitive pcb. That means we need to mount the diode at the right distance and as i understand it, let the scanning mirror write the first line of data, then shift the board by the required distance and off we go! I have investigated the diode on a hp laser printer but the frequency is wrong for the photoresist, so it will need replacing. I would like to try this but, sigh, time is lacking. Anybody interested to try? Heck it could be commercial too! Cheers happymacer

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s